(57) A video decoder, comprising
a memory (209) for storing a decoded image (210) of a previously decoded frame,
a synthesizer (1011, 1014) for synthesizing a predicted image (212) of a present frame
by using said decoded image (210) and received information about said present frame,
wherein said synthesizer (1011, 1014) includes:
assuming sampling intervals of the pixels to be 1 in both horizontal and vertical
directions with respect to the image, and assuming sampling points for which horizontal
and vertical components of coordinates of the sampling points are both integers, means
(1011, 1014) for calculating a motion vector of each pixel in said predicted image
(212) of the present frame by using motion vectors of representative points (602,
603, 604) having coordinates (i, j), (i+p, j), and (i, j+q), where i and j are integers;
means for setting the values of p and q to integer powers of 2 regardless of the size
of said predicted image (212); and
means (1011, 1014) for synthesizing said predicted image (212) by said motion vector
of each pixel.
|
|
